No abstract

With the development of machine learning, to improve the accuracy in recommendation systems, the main purpose of the suggested approach consists of using techniques and algorithms which can predict and suggest relevant tourist services (k-items) to users according to their interests, needs, or tastes. In this research, we describe how machine learning techniques can automatically provide personalized recommendations to the requestors by considering both their preferences and their implicit/explicit contextual information and the current contextual constraints of the points of interest. We build an efficient, intelligent H-RN algorithm that hybridizes both the most known machine learning algorithms, namely Random Forest and Naïve Bayes, with both collaborative filtering techniques (model-based and memory-based technique). Different experiments of our approach as part of a recommender system in the touristic field are performed over the four large real-world datasets. Recommender systems can use H-RN to improve recommendation prediction and reduce the search space of tourist services. Moreover, the results of recall, precision, accuracy, F-measure, and average rate, as well as a set of statistical tests (One-way ANOVA, Diversity) and error metrics (RMSE, MAE) have been discussed to show the improvement of the prediction accuracy of our algorithm compared to the baseline approaches in various settings.

In this paper, we present a fine-grained matching method of the services based on a hybrid similarity measure. We propose a novel encoding of the services descriptions, allowing the match between a request and an advertisement in order to make more efficient publishing and searching process of Web services and reduce the number of comparisons required. By this kind of similarity between concepts of profile, a precise matching method is developed to match the profile of the Web services and user. Searching process in the UDDI registry is done via an algorithm that allows us to extract the search concepts and retrieve the topk services, thereby further reducing the search engine's response time. The approach is illustrated through some experiments both on real and synthetic data to demonstrate its consistency and effectiveness.

No abstract

No abstract